DriftSR: One-Step Real-World Image Super-Resolution via Distribution Drifting

arXiv:2610.04819v1 Announce Type: new Abstract: One-step real-world image super-resolution (Real-ISR) offers efficient inference, but recovering realistic and perceptually rich details often relies on score distillation or adversarial learning, introducing additional trainable components and making optimization more cumbersome. To this end, we propose DriftSR, a one-step Real-ISR framework that leverages pretrained diffusion priors through distribution drifting. Specifically, we perform drifting in the frozen intermediate representation space of a pretrained diffusion model, without introducing an additional task-specific feature encoder. Building on this space, we introduce Spatial Feature Drifting, which treats spatial features rather than entire images as distributional samples, enabling denser supervision for distribution alignment. To mitigate structural deviations, we further introduce Structure-Modulated Guidance, which adaptively refines drifting guidance according to local structural consistency with the LQ input. Consequently, DriftSR optimizes only the one-step generator, without auxiliary distillation branches or adversarial discriminators. Extensive experiments on three real-world benchmarks demonstrate that DriftSR delivers high-quality super-resolution reconstruction with efficient one-step inference.

Related

• PixelIR: Fidelity-Perception Decoupling via Pixel-Space Image-Residual Flow Matching for Efficient One-Step Real-World Super-Resolution
• Degradation-Aware and Structure-Preserving Diffusion for Real-World Image Super-Resolution
• PhoenixSR: Generative Heterogeneous Distillation Unleashes Efficient Models for Real-World Super-Resolution
• VICR: Visual In-Context Restoration for Real-World Image Super-Resolution

Source: arXiv cs.CV | 2026-10-06